A novel methodology for investigating the through-thickness molten pool shape during remote laser beam welding

OPTICS AND LASER TECHNOLOGY(2024)

引用 0|浏览1
暂无评分
摘要
Molten pool shape can retrieve information on thermal gradients and solidification rates which provide substantial process signatures on cracking and weld quality. However, monitoring the molten pool shape remains a challenging task due to the complex nature of the laser welding process. The current study employs a novel monitoring methodology to examine the role of through-thickness molten pool shape in solidification cracking in partially penetrated welds. A Coherent ARM (Adjustable Ring mode) fibre laser, featuring independent control of the core and ring beam with the ratio of respective beam power specified by power ratio, was employed to study different molten pool shapes. Experimental investigations further include welding of aluminium alloy and quartz glass in butt configuration with a high-speed imaging system facing the longitudinal cross-section of the weld. The molten pool shape was evaluated using the developed image processing algorithm and the curvature analysis. Furthermore, ad-hoc Digital Image Correlation (DIC) was used to examine the thermal strain development in the through-thickness direction. Results showed that the proposed methodology could provide an accurate detection of the through-thickness molten pool shape with an improved accuracy of 94.3 % compared to the advanced prior models available in the literature, with an accuracy of 30.23 % and 42.5 %. It also revealed a tail-like feature in the molten pool's rear end, which influences the crystallisation paths and facilitates premature solidification, leading to greater tensile strains during solidification. The increments in power ratio from 0.36 to 1.5 reduced the tail-like feature and reduced the shrinkage strains in the fusion zone.
更多
查看译文
关键词
Laser welding,Through-thickness molten pool shape,Solidification cracking,Adjustable Ring Mode (ARM),Image processing,Digital Image Correlation (DIC)
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要